Re: [PATCH v6 11/41] Temporarily disable unimplemented rpi4b devices

["Philippe Mathieu-Daudé" <philmd@linaro.org>]

On 26/2/24 14:39, Peter Maydell wrote:
> On Mon, 26 Feb 2024 at 13:35, Philippe Mathieu-Daudé <philmd@linaro.org> wrote:
>>
>> On 26/2/24 01:02, Sergey Kambalin wrote:
>>> +static void raspi4_modify_dtb(const struct arm_boot_info *info, void *fdt)
>>> +{
>>> +    uint64_t ram_size;
>>> +
>>> +    /* Temporarily disable following devices until they are implemented */
>>> +    const char *nodes_to_remove[] = {
>>> +        "brcm,bcm2711-pcie",
>>> +        "brcm,bcm2711-rng200",
>>> +        "brcm,bcm2711-thermal",
>>> +        "brcm,bcm2711-genet-v5",
>>> +    };
>>> +
>>> +    for (int i = 0; i < ARRAY_SIZE(nodes_to_remove); i++) {
>>> +        const char *dev_str = nodes_to_remove[i];
>>> +
>>> +        int offset = fdt_node_offset_by_compatible(fdt, -1, dev_str);
>>> +        if (offset >= 0) {
>>> +            if (!fdt_nop_node(fdt, offset)) {
>>
>> Peter, I remember a discussion where you wre not keen on altering DTB
>> for non-Virt machines.
>>
>> Since these devices are all implemented at the end of the series, why
>> not add the devices then the raspi4 board at the end, so this patch is
>> not even required?
> 
> I'm not super-keen on it, but as you say it goes away once all
> the devices are implemented, so I'm not too worried.
> 
> Doing it this way around would let us take the first 11 patches
> in the series into git now (they've all been reviewed), which
> gives us (I think) a functional raspi4 with some missing devices,
> which seems useful in the interim until the rest of the series
> gets reviewed and committed.

Fine by me! Sergey, don't we also need patch #39 (Add missed BCM2835
properties) to have a happy Linux boot?

Patch #17 "Implement BCM2838 thermal sensor" could also go in but it
doesn't apply cleanly on top of 1-12); maybe Sergey can send a series
of "patches already reviewed" on top so they get in for v9, postponing
pcie/network for after release.

Regards,

Phil.